Search for Point-Like Tev Sources in the Large Magellanic Cloud∗

Home , SN 1987A

Search for Point-Like TeV Sources in the Large Magellanic Cloud∗

Nukri Komin† School of Physics, University of the Witwatersrand, Johannesburg, South Africa E-mail: [email protected]

Maria Haupt DESY, Zeuthen, Germany E-mail: [email protected]

for the H.E.S.S. Collaboration ‡§

The Large Magellanic Cloud (LMC) is an irregular satellite galaxy of the Milky Way, which has been observed extensively in Very-High-Energy (VHE) gamma rays with the H.E.S.S. telescopes since 2004 and reaches now a total observation time of 280 h. The exposure of the LMC is rather inhomogeneous, the region around the Tarantula Nebula having an exposure of up to 220 h while the exposure in the outer parts of the LMC is as low as 5h. A search for point-like sources was performed on this data set. This search resulted in the detection of the four already known sources (N 157B, N 132D, 30 Dor C and LMC P3) but no further signiﬁcant emission was revealed. Based on catalogues of pulsars, supernova remnants and high-mass X-ray binaries upper limits on the gamma-ray ﬂux of these objects were derived. In this talk updated results on the known gamma-ray sources as well as upper limits on the non- detected objects will be presented. It will be shown that for a large part of the LMC the existence of VHE gamma-ray sources with a similar luminosity as the already known sources can be excluded.

36th International Cosmic Ray Conference -ICRC2019- July 24th - August 1st, 2019 arXiv:1908.04656v2 [astro-ph.HE] 22 Jan 2020 Madison, WI, U.S.A.

∗The proceedings have been updated on 2020/01/22. In the original version the upper limits on the luminosity and derived values were wrongly calculated. †Speaker. ‡for collaboration list see PoS(ICRC2019)1177 §https://www.mpi-hd.mpg.de/hfm/HESS/pages/collaboration/

c Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). http://pos.sissa.it/ Search for Point-Like TeV Sources in the Large Magellanic Cloud Nukri Komin

1. Introduction

The Large Magellanic Cloud is a satellite galaxy of the Milky Way. It has a stellar mass of about 4% of that of the Milky Way [1]. The LMC is located at a distance of about 50 kpc [2] and is seen nearly face-on [2]. It harbours a number of potential gamma-ray emitting objects, among those supernova remnants (SNRs) [3], pulsars and pulsar wind nebulae (PWNe) [4, 5] and High-Mass X-Ray Binaries (HMXB) [6]. The LMC is located in the Southern Hemisphere and currently the High Energy Stereoscopic System (H.E.S.S.) is the only instrument to observe the LMC in TeV gamma rays. H.E.S.S.has observed the LMC quite extensively since its inauguration in 2004. These observations lead to the discovery of TeV gamma-ray emission from the PWN N 157B [7, 8], the superbubble 30 Dor C [8], the SNR N 132D [8] and the gamma-ray binary LMC P3 [9]. These observations clearly show the presence of TeV gamma-ray emitting sources in the LMC and the capability of H.E.S.S. to detect their emission. The work presented here searches for further TeV gamma-ray emitting sources in the LMC.

2. Data Set and Analysis

Observations of the LMC with H.E.S.S. are carried out on a nearly yearly basis. While in the beginning the focus was put on the region around the supernova remnant SN 1987A and subse- quently on the newly detected sources, in the recent years attempts were made to cover as much as possible of the entire LMC. Up to know an exposure of 280 h has been recorded. The left panel of Fig. 1 shows the exposure contours overlaid on an optical sky map. It can be seen that the LMC is rather inhomogeneously covered, with about 10 h of exposure on the western edge and more than 200 h around the Tarantula nebula and the already detected PWN N 157B. The data were analysed using Model analysis [11], where the camera images are compared with simulations using a log-likelihood minimisation. In addition to the standard cuts an optional cut on the reconstructed direction error was applied in order to increase the angular resolution

Figure 1: Left panel: Optical image of the LMC (from [10]) with overlaid H.E.S.S. exposure contours. The contour lines denote (from outside to inside) 10, 20, 50, 100 and 200 h effective exposure time. Right panel: Signiﬁcance map of the H.E.S.S. data set with overlaid exposure contours.

1 Search for Point-Like TeV Sources in the Large Magellanic Cloud Nukri Komin and further decrease the background. A similar cut was used in previous publications [8, 9]. The background was estimated from rings around each sky position to generate the gamma-ray image [12]. The energy threshold for this data set is 714 GeV. The right panel of Fig. 1 shows the signiﬁcance map for this data set. It is the ﬁrst time that a complete view of the LMC in TeV gamma rays is presented.

3. The Pulsar Wind Nebula N 157B

In the signiﬁcance map in Fig. 1 the ) emission from the PWN N 157B can be -1 H.E.S.S. (preliminary) clearly seen. With a ﬂux of about 2% of TeV

8− -1 12−

s 10

the ﬂux of the Crab Nebula N 157B is by far -2

the brightest VHE gamma ray source in the (cm 2

LMC. It is detected in yearly subsets of the E × Flux

8.5− deep data set of the LMC and can therefore be used as a standard candle for the LMC. For the study of this particular object

13 9− the additional high-resolution cut was not ap- 10 1 10

0 0.2 0.4 0.6 0.8 1 1.2

) χ 4 1 10 plied, in order to obtain more excess counts. 3 Residuals ( Δ Residuals In the current data set 1152 excess events 2 with a statistical signiﬁcance of 28.6σ are 1 0 detected from N 157B. Figure 2 shows the 1− gamma-ray energy spectrum of this object. 2− The spectrum can be ﬁt with a curved power 3−

4− 0 0.2 0.4 0.6 0.8 1 1.2 law 1 10 Energy (TeV) −α−β×ln( E ) dN E 1.5TeV Figure 2: Gamma-ray energy spectrum of the PWN = Φ0 , dE 1.5TeV N 157B. (3.1) −9 −2 −1 −1 with Φ0 = (3.6 ± 0.1)×10 m s TeV , α = (2.23 ± 0.08) and β = (0.32 ± 0.07) . This three-parameter model is preferred by 4.4σ over a simple power law. The gamma-ray luminosity in the 1 − 10 TeV energy range for a distance of 50 kpc is (6.5 ± 0.3) × 1035 erg/s and matches within errors the previously reported value [8].

4. Upper Limits on Gamma-Ray Fluxes from Selected Sources

The data set is searched for emission from pulsars, SNRs and HMXBs. For each of the objects found in the catalogues a dedicated analysis was performed. No significant emission was found from any of the objects. The left panel of Fig. 3 shows the significance distribution which can be fit by a Gaussian with a mean of 0.4 and a standard deviation of 1.2. All known sources have been removed from this plot. The only outlying object is SNR B0536−6914 with a statistical significance of 6.6σ. A closer inspection of the object shows that this is spill-over emission from N 157B and 30 Dor C and not genuine emission from this object itself.

2 Search for Point-Like TeV Sources in the Large Magellanic Cloud Nukri Komin

20.0 H.E.S.S. (preliminary) 10 8 H.E.S.S. (preliminary) N 157B 17.5

15.0

12.5 LMC P3 30 Dor C 10.0 10 9 N 132D 7.5 F (>1 TeV) [m-2 s-1]

5.0 pulsar 2.5 SNR HMXB 0.0 2 0 2 4 6 100 101 102 significance live time [h]

Figure 3: Left panel: Significance distribution of the all objects (excluding the detected sources). Right panel: Upper limits on the gamma-ray flux from pulsars, SNRs and HMXB depending on the exposure time of the observations. The red line indicates the limit of 20 h of exposure time. The green line indicate the flux levels of the already detected sources. These limits are taken from [8] for N 157B, 30 Dor Cand N 132D and from [9] for LMC P3.

Upper limits on the gamma-ray photon flux between 1 and 10 TeV for a confidence level of 95% and an assumed spectral index of -2.3 are computed for all objects. The right panel of Fig. 3 shows the derived upper limits depending on the corresponding exposure time for each object. Green lines denote flux levels of detected sources. Further sources with flux levels similar to N 157B can be excluded. Where the exposure time exceeds 20 h it can also be excluded that sources exhibit a gamma-ray flux at the level of 30 Dor C, N 132D or LMC P3. It can be seen that for more than 20 h exposure only marginal improvement of the upper limits can be achieved with increasing exposure time. Therefore, an increase of exposure time beyond 20 h will not increase the detection capability of sources in the LMC. On the other side, the detection of new sources is still possible by increasing the exposure time to 20 h everywhere in the LMC. In the following only upper limits on objects with a minimum exposure of at least 20 h are reported.

4.1 Pulsars and Pulsar Wind Nebulae The loss of rotational energy of pulsars can power PWNe. These nebulae are the most abundant emitters of galactic TeV emission (see [13, 14]). Further on, pulsed gamma-ray emission in the 100 GeV to TeV range has been detected from a few pulsars themselves: the Crab pulsar [15, 16], the Vela pulsar [17] and most recently from Geminga. These findings motivate to search for TeV gamma-ray emission near pulsars in the LMC. The H.E.S.S. LMC data set has been analysed at source positions corresponding to the pulsars found in the ATNF pulsar data base [4] and in [5]. Apart from N 157B (see above) no significant emission from any other object was detected. The significances and upper limits on the photon flux are reported in Table 1. The photon flux limits between 1 and 10 TeV are below 10−9m−2 s−1, well below the photon flux from N 157B of 7.5 ×10−9m−2 s−1. The upper limit on the gamma-ray luminosity, L, can be related to the spin-down luminosity, E˙, of the pulsar, resulting in an upper limit on the pulsar efficiency. These efficiencies are typically of the order of one percent [18], it

3 Search for Point-Like TeV Sources in the Large Magellanic Cloud Nukri Komin

Table 1: Results from the analysis of pulsars in the LMC. The upper limit on the luminosity is calculated for an assumed distance of 50 kpc. The spin-down luminosities are taken from ATNF [4], a dash indicates no available value.

name Sig UL L Edot PSR eff s−1 m−2 ergs−1 ergs−1 J0519-6932 1.2 6.9e-10 7.5e+34 1.5e+33 5.0e+01 J0521-68 1.1 4.1e-10 4.5e+34 - - J0522-6847 1.1 6.1e-10 6.6e+34 2.3e+33 2.9e+01 J0529-6652 -0.1 5.9e-10 6.4e+34 6.6e+32 9.8e+01 J0532-6639 -1.2 2.4e-10 2.6e+34 8.0e+32 3.3e+01 J0532-69 0.2 6.7e-10 7.3e+34 - - J0534-6703 -0.3 4.8e-10 5.3e+34 2.8e+33 1.9e+01 J0535-66 0.9 6.5e-10 7.0e+34 - - J0535-6935 -0.1 3.9e-10 4.3e+34 5.6e+34 7.6e-01 J0537-69 1.9 7.1e-10 7.7e+34 - - J0540-6919 2.5 6.5e-10 7.1e+34 1.5e+38 4.9e-04 J0542-68 1.0 3.0e-10 3.3e+34 - - J0543-6851 -0.7 2.5e-10 2.7e+34 4.4e+32 6.2e+01 J0555-7056 -0.2 7.7e-10 8.4e+34 4.1e+32 2.0e+02

is 0.08% for N 157B [7]. The upper limits on the efﬁciencies in Table 1 are not constraining for most of the pulsars in the LMC. The most remarkable object in this list is PSR J0540−6919, the pulsar with the third highest spin-down luminosity and for which gamma-ray pulsations have been detected by Fermi-LAT [19]. If this pulsar is a TeV gamma-ray emitter, its efﬁciency is extremely low. All other pulsars are extremely old, several hundred kiloyears, so gamma-ray emission is not necessarily expected from these objects.

4.2 Supernova Remnants

SNRs are the prime candidates for the acceleration sites of the galactic Cosmic Rays. TeV emission has been detected from a number of SNRs in the Milky Way and from N 132D in the LMC. This motivates the search for gamma-ray emission from SNRs in the LMC. Here the cata- logue of SNRs in the LMC provided by [3] is used. Three of the SNRs show significant emission: The already known N 132D and DEM L241. The emission at the position of the latter is variable and periodic and can be attributed to the binary LMC P3 inside the SNR [9]. The emission at the position of SNR B0536−6914 is due to spill-over from neighbouring sources. These three objects are removed from the list. The analysis of SNR SN 1987A does suffer from spill-over as well, but to a much lesser degree than SNR B0536−6914. The upper limits on SN 1987A are therefore very conservative. Table 2 lists the upper limits on the photon flux and luminosity of the remaining objects. The upper limit on the photon fluxes are of the order of the flux of N 132D. It can therefore be excluded that any of these SNRs emits TeV gamma-ray emission similar to N 132D. The corresponding total energy in protons is calculated for an assumed ambient density of n = 1cm−3 with a cooling time

4 Search for Point-Like TeV Sources in the Large Magellanic Cloud Nukri Komin

Table 2: Results from the analysis of SNRs in the LMC. The upper limit on the luminosity is calculated for an assumed distance of 50 kpc. The upper limit on the total energy is calculated 15 with a cooling time of τpp = 4×10 s assuming Table 3: Upper limits on high-mass X-ray − an ambient target density of 1cm 3. binaries in the LMC.

name Sig UL L Wtot name Sig UL L s−1 m−2 ergs−1 erg s−1 m−2 ergs−1 1RXSJ053353.6-7204 1.3 1.0e-09 1.1e+35 4.3e+50 1A 0535-668 1.1 6.4e-10 7.0e+34 B0509-67.5 1.0 1.1e-09 1.2e+35 4.7e+50 1SAX J0544.1-7100 0.2 3.9e-10 4.2e+34 B0519-690 -0.2 4.7e-10 5.1e+34 2.0e+50 2A 0532-664 -0.2 1.9e-10 2.1e+34 B0520-694 1.4 8.0e-10 8.7e+34 3.5e+50 3A 0540-697 0.7 6.1e-10 6.6e+34 B0528-6716 -1.3 2.6e-10 2.8e+34 1.1e+50 H 0544-665 -0.3 3.5e-10 3.8e+34 B0528-692 -2.6 1.5e-10 1.7e+34 6.7e+49 RX J0456.9-6824 0.6 7.3e-10 7.9e+34 B0532-675 -0.5 1.8e-10 2.0e+34 7.9e+49 RX J0507.6-6847 -1.3 2.7e-10 3.0e+34 B0534-699 -1.1 5.3e-10 5.8e+34 2.3e+50 RX J0512.6-6717 1.9 1.2e-09 1.4e+35 B0536-6914 7.0 1.5e-09 1.6e+35 6.4e+50 RX J0516.0-6916 0.4 5.9e-10 6.4e+34 B0540-693 2.6 6.7e-10 7.3e+34 2.9e+50 RX J0520.5-6932 1.9 8.3e-10 9.0e+34 B0548-704 -0.0 3.6e-10 4.0e+34 1.6e+50 RX J0523.2-7004 0.2 5.6e-10 6.2e+34 DEM L109 -0.7 3.5e-10 3.8e+34 1.5e+50 RX J0524.2-6620 -2.2 3.4e-10 3.7e+34 DEM L175a -0.0 8.9e-10 9.7e+34 3.9e+50 RX J0527.1-7005 1.0 5.8e-10 6.3e+34 DEM L205 0.5 2.8e-10 3.1e+34 1.2e+50 RX J0529.4-6952 0.4 5.0e-10 5.4e+34 DEM L214 0.0 5.0e-10 5.5e+34 2.2e+50 RX J0532.3-7107 -0.5 3.4e-10 3.7e+34 DEM L218 -1.3 2.4e-10 2.6e+34 1.0e+50 RX J0535.0-6700 -0.7 3.5e-10 3.8e+34 DEM L238 -0.5 3.8e-10 4.2e+34 1.7e+50 RX J0535.6-6651 1.3 6.9e-10 7.5e+34 DEM L249 1.1 5.7e-10 6.2e+34 2.5e+50 RX J0541.4-6936 -0.9 3.3e-10 3.6e+34 DEM L256 -0.3 4.2e-10 4.6e+34 1.8e+50 RX J0541.5-6833 -1.4 1.3e-10 1.4e+34 DEM L299 2.7 5.6e-10 6.1e+34 2.4e+50 RX J0546.8-6851 0.1 1.7e-10 1.9e+34 DEM L316A -1.4 1.4e-10 1.5e+34 5.9e+49 XMMU J053115.4-705350 0.8 5.1e-10 5.6e+34 DEM L316B -2.0 1.1e-10 1.2e+34 4.7e+49 XMMU J054134.7-682550 -0.7 1.4e-10 1.6e+34 DEM L71 0.7 8.7e-10 9.5e+34 3.8e+50 Honeycomb 0.2 4.3e-10 4.7e+34 1.9e+50 J0508-6830 -0.1 4.7e-10 5.1e+34 2.1e+50 J0511-6759 2.1 1.1e-09 1.2e+35 4.8e+50 J0514-6840 2.0 9.5e-10 1.0e+35 4.2e+50 J0517-6759 2.2 6.7e-10 7.3e+34 2.9e+50 J0550-6823 0.1 2.3e-10 2.5e+34 1.0e+50 N103B 1.8 9.2e-10 1.0e+35 4.0e+50 N120 -0.5 4.9e-10 5.3e+34 2.1e+50 N159 -0.6 3.6e-10 4.0e+34 1.6e+50 N206 1.4 8.8e-10 9.6e+34 3.8e+50 N23 0.9 7.8e-10 8.5e+34 3.4e+50 N44 1.1 4.2e-10 4.6e+34 1.8e+50 SNR1987A 1.5 5.9e-10 6.4e+34 2.6e+50 of n −1 τ = 4 × 1015 s. (4.1) pp 1cm−3 It can be seen from Table 2 that the total energy in protons roughly corresponds to the canonical 10% of the explosion energy of 1051erg. It should be noted that the gamma-ray upper limits are for the 1–to–10 TeV energy range and therefore probing protons above ∼ 10TeV only. Any lower energy protons are not considered in the limits presented here. Further on, a higher ambient target density would lower the required total energy in protons.

4.3 High-Mass X-Ray Binaries Gamma-ray binaries are a growing class of HMXB systems exhibiting strong emission in the gamma-ray energy regime. The latest discovery is the binary system LMC P3 in the LMC [9, and references therein]. This motivates a search for TeV gamma-ray emission from other HMXBs in the LMC. The positions of the HMXBs in [6] are analysed. No significant emission from any of these objects is found and upper limits on the photon flux are shown in Table 3. The upper limit on the photon fluxes are of the order of the flux of LMC P3. It can therefore be excluded that any of

5 Search for Point-Like TeV Sources in the Large Magellanic Cloud Nukri Komin these binaries emits TeV gamma-ray emission similar to LMC P3. However, this study considers only the average ﬂux over the entire orbit. LMC P3 emits only during 20% or less of its orbit. There for a search for pulsed emission from the binary systems still has discovery potential. The upper limit on the luminosity of the HMXBs can be compared with the accretion luminosity of stellar wind from the companion star onto a hypothetical pulsar, ˙ M Mco 10km 36 −1 Lacc = −10 × 1.2 × 10 ergs , (4.2) 10 M /yr 1.4M Rco where Mco and Rco are the mass and radius of the compact object [20]. M˙ is the mass-accretion rate −10 which is of the order of 10 M /yr in a typical close binary system with accretion of the stellar wind. The upper limits on the luminosities of the HMXBs corresponds to a few percent of this accretion efﬁciency. If accretion is taking place in these systems then the conversion of accretion luminosity into TeV gamma-rays is lower then a few percent. However, a much higher accretion rate during only a part of the orbit cannot be ruled out.

5. Conclusion

The H.E.S.S. Collaboration is expanding the observations of the LMC in order to achieve a full survey of the galaxy. It is shown that a minimum exposure of 20 h is sufficient to test for emission at the level of the already detected sources. The search for emission from pulsars, SNRs and HMXB did not reveal any new sources. Upper limits on these objects are presented where a minimum exposure of 20 h of exposure was achieved. These upper limits probe to some extent the physical parameter space of gamma-ray production in these objects. The increased exposure on the PWN N 157B shows a clear curvature of the spectrum. More detailed modelling of the emission including the measurements at lower energies carried by Fermi- LAT will be performed. The work presented here defines the baseline for the continued observations with H.E.S.S. as well as future observations with the Cherenkov Telescope Array (CTA). With the H.E.S.S. telescopes only emission at flux levels of the already detected sources can be measured. CTA will be able to probe for much fainter objects [21].

Acknowledgements

The support of the Namibian authorities and of the University of Namibia in facilitating the construction and operation of H.E.S.S. is gratefully acknowledged, as is the support by the German Ministry for Education and Research (BMBF), the Max Planck Society, the German Research Foundation (DFG), the Helmholtz As- sociation, the Alexander von Humboldt Foundation, the French Ministry of Higher Education, Research and Innovation, the Centre National de la Recherche Scientiﬁque (CNRS/IN2P3 and CNRS/INSU), the Com- missariat Ãa˘ l’Ãl’nergie atomique et aux Ãl’nergies alternatives (CEA), the U.K. Science and Technology Facilities Council (STFC), the Knut and Alice Wallenberg Foundation, the National Science Centre, Poland grant no. 2016/22/M/ST9/00382, the South African Department of Science and Technology and National Research Foundation, the University of Namibia, the National Commission on Research, Science & Tech- nology of Namibia (NCRST), the Austrian Federal Ministry of Education, Science and Research and the Austrian Science Fund (FWF), the Australian Research Council (ARC), the Japan Society for the Promotion of Science and by the University of Amsterdam.

6 Search for Point-Like TeV Sources in the Large Magellanic Cloud Nukri Komin

We appreciate the excellent work of the technical support staff in Berlin, Zeuthen, Heidelberg, Palaiseau, Paris, Saclay, TÃijbingen and in Namibia in the construction and operation of the equipment. This work beneﬁtted from services provided by the H.E.S.S. Virtual Organisation, supported by the national resource providers of the EGI Federation.

References

[1] P. J. McMillan. In: MNRAS 414 (July 2011), pp. 2446–2457. [2] G. PietrzyÅDskiˇ et al. In: Nature 495 (Mar. 2013), pp. 76–79. [3] P. Maggi et al. In: A&A 585, A162 (Jan. 2016), A162. [4] R. N. Manchester et al. In: AJ 129 (Apr. 2005), pp. 1993–2006.

[5] J. P. Ridley et al. In: MNRAS 433.1 (May 2013), pp. 138–146. ISSN: 0035-8711. [6] Q. Z. Liu, J. van Paradijs, and E. P. J. van den Heuvel. In: A&A 442.3 (Nov. 2005), pp. 1135– 1138. [7] H. E. S. S. Collaboration. In: A&A 545, L2 (Sept. 2012), p. L2. [8] H. E. S. S. Collaboration. In: Science 347.6220 (Jan. 2015), pp. 406–412. [9] HESS Collaboration. In: A&A 610, L17 (Mar. 2018), p. L17. [10] A. Mellinger. In: PASP 121 (Nov. 2009), p. 1180. [11] Mathieu de Naurois and LoÄsÌ´ ´Lc Rolland. In: Astroparticle Physics 32.5 (Dec. 2009), pp. 231–252. [12] D. Berge, S. Funk, and J. Hinton. In: A&A 466.3 (May 2007), pp. 1219–1229. [13] H. E. S. S. Collaboration. In: A&A 612, A1 (Apr. 2018), A1. [14] H. E. S. S. Collaboration. In: A&A 612, A2 (Apr. 2018), A2. [15] VERITAS Collaboration. In: Science 334.6052 (Oct. 2011), p. 69. [16] MAGIC Collaboration. In: A&A 585, A133 (Jan. 2016), A133. [17] H. E. S. S. Collaboration. In: A&A 620, A66 (Dec. 2018), A66. [18] E. de OÃsa´ Wilhelmi. In: High-Energy Emission from Pulsars and their Systems. Ed. by D. F. Torres and N. Rea. 2011, pp. 435–452.

[19] Fermi/LAT Collaboration. In: Science 350.6262 (2015), pp. 801–805. ISSN: 0036-8075. [20] J. Frank, A. King, and D. Raine. 1992. [21] P. Martin et al. In: Science with the Cherenkov Telescope Array. Edited by CTA Consortium. Published by World Scientiﬁc Publishing Co. Pte. Ltd., 2019. ISBN #9789813270091, pp. 125-141. 2019, pp. 125–141.